Human Reproduction vol.14 no.6 pp.

16421646, 1999

Human chorionic gonadotrophin concentrations in early

pregnancy after in-vitro fertilization

Sverre Bjercke1,3, Tom Tanbo1, Per Olav Dale,1

Lars Mrkrid2 and Thomas byholm1
1Department

of Obstetrics and Gynecology and 2Department of

Clinical Chemistry, Rikshospitalet, 0860 Oslo, Norway

3To

whom correspondence should be addressed

There is increased risk of early pregnancy loss after

assisted reproduction. In this study the use of serum human
chorionic gonadotrophin (HCG) concentrations on day 12
after in-vitro fertilization (IVF) and embryo transfer was
evaluated to predict pregnancy outcome. A total of 417
IVF pregnancies were included. Early pregnancy loss was
defined as biochemical pregnancies, ectopic pregnancies
and first trimester abortions. Vital pregnancies were defined
as delivered singletons, multiple pregnancies and second
trimester abortions. On the post embryo transfer day 12,
the mean HCG concentration of the vital pregnancy group
was significantly higher than in early pregnancy loss outcomes (P < 0.00001). Receiver operating characteristic
(ROC) curve analysis was performed to evaluate the cutoff value of HCG giving maximal sensitivity and specificity
in order to discriminate early pregnancy losses from vital
pregnancies. A patient with a HCG value higher than the
calculated cut-off value (55 IU/l) had a 90% chance of
having a vital pregnancy after IVF and embryo transfer.
It can be concluded that a discriminatory HCG value on
day 12 after IVF and embryo transfer cycles may be useful
in predicting pregnancy outcome and may guide clinicians
in identifying those pregnancies at risk for adverse outcomes and instituting more intensive surveillance in this
population.
Key words: clinical pregnancy/human chorionic gonadotrophin/
in-vitro fertilization

Introduction
Pregnancies obtained after in-vitro fertilization (IVF) and
embryo transfer are at increased risk of adverse outcome
compared with natural conceptions (Edwards et al., 1984;
Ben-Rafael et al., 1988; Correy et al., 1988). A reliable
and inexpensive diagnostic test to differentiate between vital
pregnancies and pregnancies with early adverse outcome might
reduce the psychological tension and anxiety present in many
of these patients, and also reduce the cost by making the
treatment more efficient. On the other hand, in patients with
a high risk of unfavourable outcome based on the test a more
careful follow up might reduce the risks associated with these
abnormal pregnancies.
1642

Ultrasound (US) examination is effective for evaluation of

ongoing pregnancies, but a gestational sac is not reliably
visible until 3337 days after the luteinizing hormone (LH)
surge (Shapiro et al., 1992). As a result of this inability of US
to identify very early pregnancy abnormalities, there is an
ongoing effort to institute a method that can forecast pregnancy
outcome. Several studies have investigated hormones like
progesterone, oestradiol, relaxin, pregnancy-specific 1-glycoprotein, inhibin, and CA-125 in predicting prognosis (Witt
et al., 1990; Hahlin et al., 1991). The exact role of these
markers and their clinical utility, however, has not yet been
established. Researchers have reported the usefulness of early
serum human chorionic gonadotrophin (HCG) (Confino et al.,
1986; Dor et al., 1988; Heiner et al., 1992; Fridstrm et al.,
1995; Glatstein et al., 1995). One of these studies has reported
clinically discriminatory HCG values on different days post
embryo transfer (Glatstein et al., 1995), but none of them as
early as on day 12.
The aim of this study was to investigate whether a single
HCG value on day 12 after embryo transfer could be used as
a marker to predict the outcome of a pregnancy after IVF.
Materials and methods
Subjects
During the period of January 1994 to December 1996, 417 pregnancies
obtained after IVF and embryo transfer treatment for infertility due to
tubal factor (61%), polycystic ovarian syndrome (4%), endometriosis
(12%), unexplained (14%), or male factor (9%) were studied. Conception was defined as a HCG serum concentration higher than 10 IU/l.
All patients had their HCG assay performed inside our institution on
day 12 after embryo transfer. The median age of the participant was
33 (2144 years).
In-vitro fertilization protocol
The patients underwent IVF and embryo transfer according to
described protocols (byholm et al., 1990; Tanbo et al., 1995).
Pituitary down regulation with a gonadotrophin releasing hormone
agonist (GnRHa) (Suprefact; Hoechest, Frankfurt am Main,
Germany) was used in all patients. Ovarian hyperstimulation was
performed using human menopausal gonadotrophin (HMG)
(Pergonal; Serono, Aubonne, Switzerland) and follicle stimulating
hormone (FSH) (Fertinorm and Fertinorm HP; Serono). Transvaginal ultrasound (US) and serum oestradiol concentrations were used
to monitor the IVF cycle. When three or more of the follicles had a
diameter exceeding 18 mm, 10 000 IU units of HCG (Profasi;
Serono) were administered s.c. to induce ovulation. Oocyte aspiration
was performed approximately 3436 h after HCG was given using a
standard transvaginal US-guided approach. Embryo transfer was
performed 2 days after oocyte aspiration and a maximum of two
embryos was transferred. All patients included in this series received
European Society of Human Reproduction and Embryology

HCG concentration in predicting pregnancy outcome

progesterone support for 4 weeks, either as vaginal suppositories

(600 mg daily) (Progestan; Organon, Oss, The Netherlands) or i.m.
injections (25 mg daily) starting on the day prior to oocyte aspiration.
Serum HCG assay
Samples of venous blood were collected routinely on the 12th day
after embryo transfer. Serum HCG concentrations were measured by
microparticle enzyme immunoassay for the intact HCG molecule
(Imx; Abbot Laboratories, Abbot Park, IL, USA) at the Central
Chemical Laboratory of our hospital. The assay was calibrated using
the World Health Organization (WHO) Third International Standard
(75/537) (formerly designated the WHO First International Reference
Preparation). The intra-assay coefficient of variation (CV) was 4.1%
for a mean HCG of 23 IU/l; 3.9% for a HCG mean of 143 IU/l, and
3.4% for a mean HCG of 707 IU/l. The inter-assay CV was 4.6%,
6.1% and 9.6% respectively, and the sensitivity was 2 IU/l. Serum
samples that were not obtained by day 12 after embryo transfer were
discarded. As the HCG analyses were only performed on Monday,
Wednesday and Friday, and not at weekends, some of the samples
were stored for 1 to 2 days before being tested. If stored for more
than 24 h, the samples were frozen at 24C. Otherwise they were
maintained at 4C.
Pregnancy outcome
Pregnancy outcomes were divided into two groups: early pregnancy
losses and vital pregnancies. Early losses included pregnancies
classified as chemical (subclinical), ectopic gestations and first trimester spontaneous abortions. Vital pregnancies were defined as delivered
singletons, multiple pregnancies or second trimester abortions occurring with a documented fetal heart, before the abortion. Our series
contained five patients who experienced ectopic pregnancy. No
heterotopic pregnancies were detected among our patients during the
study period. The same study was performed with ectopic pregnancies
excluded from the failures, and only singleton pregnancies were
included in the successful group.
Statistical analysis
The MannWhitney U test was used to compare the differences
between the groups. A receiver operating characteristic (ROC) curve
analysis (McNeil et al., 1975; Hanley and McNeil, 1982) was used
to estimate the predictive power of the measured variables (Swets,
1979), and to determine the HCG cut-off that optimally discriminated
early pregnancy losses from vital pregnancy cycles. ROC curve was
obtained by plotting the sensitivity of the HCG value versus 1specificity for a series of multiple cut-off points on the 12th post
embryo transfer day.
The cut-off point was chosen to maximize sensitivity as well as
specificity. As a measure for the diagnostic ability of the cut-off to
predict pregnancy outcome, the positive and negative predictive
values were determined at the calculated cut-off value. As described
(Glatstein et al., 1995) the following definitions were used: sensitivity:
the probability that a patient with an early pregnancy loss will have
an HCG value less than the cut-off value, the true positive rate;
specificity: the probability that a patient with a vital pregnancy will
have HCG value greater than the cut-off value; positive predictive
value: the probability that a patient with a HCG value less than the
cut-off will have an early pregnancy loss; and negative predictive
value: the probability that a patient with an HCG value greater than
the cut-off will have a vital pregnancy. Then positive and negative
post test probability were calculated and found to coincide with the
positive and negative predictive value respectively. Unless otherwise
stated, values are expressed as means 6 SE of the mean with a
probability value of P , 0.05.

Table I. Pregnancy outcome in patients undergoing in-vitro fertilization

with an initial human chorionic gonadotrophin (HCG) value .10 mIU/l
Pregnancy outcome

n (%)

Total no. of pregnancies

Early pregnancy losses:
biochemical pregnancies and
spontaneous abortions (first trimester)a
ectopic pregnancies
Vital pregnancies:
delivered singleton
delivered twins
delivered triplets
spontaneous abortion (second trimester)

417
107 (26)

aIncluding

102
5
310
239
67
3
1

(24)
(1)
(74)
(57)
(16)
(1)

chemical abortions.

Figure 1. Day 12 post embryo transfer (mean 6 SEM) of serum

human chorionic gonadotrophin (HCG) in spontaneous abortions
(sp. abort) (n 5 107), viable singletons (n 5 239) and viable twins
(n 5 67). The receiver operating characteristic (ROC) calculated
cut-off is shown in a hatched line, dividing the spontaneous
abortions from viable pregnancies.

Results
Table I shows the pregnancy outcomes of the 417 studied
patients. Vital pregnancies represented 74% and early pregnancy failure 26% of all pregnancies, respectively. There is a
statistically significant difference between the HCG concentrations in the early pregnancy failure group versus the vital
pregnancy group (P , 0.00001). The HCG concentration
(mean 1 SE) for the three groups: spontaneous abortions, vital
singletons and vital twins are illustrated in Figure 1. Using
the MannWhitney U test, there was a statistically significant
difference in serum HCG values between singleton and multiple
pregnancies (P , 0.0001) and between singletons and spontaneous abortions (P , 0.0001).
To determine the clinical significance of these statistical
findings, a ROC curve was derived (Figure 2). By analysing
the ROC curve the cut-off point giving optimal sensitivity and
specificity was found to be 55 IU/l. At the cut-off value of 55
IU/l the negative predictive value was 90%. Stated alternatively,
this means that a patient who had a HCG concentration higher
than the cut-off value had a 90% chance of having a vital
pregnancy. The positive predictive value was 60%. Therefore,
a patient in this study population who presented with a HCG
concentration less than the cut-off (55 IU/l) had a probability
of an early pregnancy failure equal to that percentage day 12
post embryo transfer. The negative and positive predictive
1643

S.Bjercke et al.

Figure 3. Horizontal bars illustrate the percentage (%) distribution

of pregnancies classified as early pregnancy losses (true positive)
and vital pregnancies (false positive) with increasing cut-off values
of human chorionic gonadotrophin (HCG). The vertical line in the
middle of the figure represents the division between the percentage
of early pregnancy losses (left) and the percentage of vital
pregnancies (right).

Figure 2. Receiver operating characteristic (ROC) curve plotting

true positive versus false positive rate and the optimal
discriminatory human chorionic gonadotrophin (HCG) cut-off for
day 12 post embryo transfer.

values were found to correspond to the negative and positive

post test probability figures which were calculated to be 91%
and 61% respectively.
When the same studies were performed excluding ectopic
pregnancies from the early pregnancy failure group, and
multiple gestations from the successful pregnancy group, an
identical cut-off value of 55 IU/l was found. Also the positive
and negative predictive values were of the same order, 61%
and 87% respectively. Thus the optimal cut-off value in order
to discriminate between spontaneous abortion and singleton
pregnancies is the same as the cut-off value that discriminates
between the groups when the ectopic pregnancies and multiple
gestations were included (55 IU/l).
Figure 3 illustrates the relative distribution of early pregnancy losses and vital pregnancies from the selected cut-off
point of 55 IU/l to 100 IU/l. At the horizontal bar representing
the chosen cut-off value at 55 IU/l 60% will have early
pregnancy losses while 40% will have vital pregnancies. With
increasing HCG cut-off values the figure shows that the relative
distribution of vital pregnancies increases.
Discussion
Various studies have addressed the question about early HCG
values and their relationship to pregnancy outcome after IVF
(Confino et al., 1986; Deutinger et al, 1986; Dor et al., 1988;
Yamashita et al., 1989; Lower and Yovich, 1993; Fridstrm
et al., 1995; Glatstein et al., 1995). This study is unique in
that the cut-off value of HCG was calculated on one single
day and as early as day 12 post embryo transfer based on a
generated ROC curve. The testing for HCG was performed at
a single laboratory only, minimizing intra-assay and inter1644

assay variability (Confino et al., 1986). This report also

included 417 IVF pregnancy cycles, the largest series to date
in the literature of this question. In addition, the values reported
were based on a commonly accepted HCG reference, the WHO
Third International Standard.
The present study shows that, by means of a single serum
HCG determination on day 12 after embryo transfer, a group
of patients can be identified that is at increased risk of carrying
an abnormal pregnancy. At the cut-off value of 55 IU/l patients
testing positively for HCG had a 61% risk of carrying an
ectopic or non-viable intrauterine pregnancy. Therefore close
monitoring of the course of the pregnancy is called for. Among
those with a HCG value higher than 55 IU/l there was a 90%
probability of a vital pregnancy. In total 26% of the patients
in this study experienced early pregnancy loss while 74% had
vital pregnancies. The latter group comprised one second
trimester loss (16th week). Second trimester abortions are in
this study regarded as successful implantation. As explained
(Glatstein et al., 1995) early HCG concentrations can predict
implantation outcome, but it cannot invariably predict pregnancy outcome. Therefore, it would be unlikely that a single
early HCG value could reliably predict the occurrence of a
live birth, an event influenced by multiple factors other than
implantation quality like a second or third trimester abortion
due to an incompetent cervix.
The calculated cut-off reported in this study was chosen to
best discriminate between early pregnancy losses and vital
pregnancies. There is a limit to what can be achieved with
biochemical diagnosis in early pregnancy, since pathological
pregnancies often have a normal growth rate early in their
course (e.g. blighted ovum, initially beating heart). The earlier
the test is performed, the higher is the rate of false negative
tests. By definition, there is a trade off between sensitivity and
specificity. If one wishes to increase the HCG cut-off value
on day 12 after embryo transfer from 55 IU/l to 100 IU/l there
is a decline in positive predictive value (true spontaneous
abortion) from 60% to 32% (Figure 3). At the same time the
sensitivity increases from 73% to 90%, while the specificity

HCG concentration in predicting pregnancy outcome

falls from 83% to 35%. Among the 417 patients studied, there
were only five (1.2%) who experienced ectopic pregnancy.
Since there were very few ectopic pregnancies in the study,
the main aim was to find a cut-off value that could discriminate
between ordinary spontaneous and successful pregnancies. It
is doubtful if any higher sensitivity (73%) than the one
estimated at the cut-off value of 55 IU/l would be more
suitable to detect ectopic pregnancies with enough certainty
to be diagnostic. Mol et al. (1997) have reported on optimal
HCG cut-off values for ectopic pregnancies in a group of 86
women where 24 (28%) experienced ectopic pregnancies (Mol
et al., 1997). Their measurements were done on days 6, 9 and
15. The authors concluded that serum HCG measurement 9
days after embryo transfer could identify pregnancy failures
with a 100% specificity at a cut-off value of 18 IU/l, but it
could not identify patients with ectopic pregnancies with
enough certainty to justify immediate treatment. Another recent
study reported no difference in the measurable isoforms of
HCG between normal intrauterine pregnancy and ectopic
pregnancies (Mock et al., 1998). Abdominal pain or bleeding
is usually an additional sign necessary to raise suspicion of
ectopic pregnancy. Under these circumstances transvaginal
sonography is of great diagnostic aid (Mol et al., 1997).
If one was to select one group for closer follow up on basis
of the HCG measured on day 12 after embryo transfer it ought
to be the patients with values equal to or below 55 IU/l. The
lower the HCG value (lower sensitivity) the higher is the
likelihood of non-viable pregnancy (specificity).
To set a higher cut-off point in order to increase the
sensitivity does not seem reasonable in as much as this would
decrease the specificity considerably and not contribute to a
better discrimination between vital and the potentially lifethreatening ectopic pregnancies.
The HCG concentrations were significantly higher on day
12 after embryo transfer for vital pregnancy to early pregnancy
losses. This confirms previous studies (Heiner et al., 1992;
Keith et al., 1993; Fridstrm et al., 1995; Glatstein et al.,
1995). By choosing day 12 as the test day, unnecessary delay
of pregnancy diagnosis could be avoided, which is crucial
both from a psychological and a medical point of view. Relief
of tension might not only be important for the patients wellbeing (Friedman, 1989), but also for the outcome of pregnancy
(Stray-Pedersen and Stray-Pedersen, 1984). For patients using
progesterone as luteal support, an early and reliable test for
pregnancy is desirable in order to terminate luteal support in
patients not achieving pregnancy.
It was shown in this series that multiple pregnancies had
significantly higher HCG values compared with singletons on
day 12 after embryo transfer. This finding is in agreement with
previous reports (Heiner et al., 1992; Fridstrm et al., 1995;
Glatstein et al., 1995) but ultrasound confirmation of the
number of fetuses and their viability still should be used to
make clinical recommendations to patients in any case of
multiple pregnancy.
Further it was calculated that this series of singletons had
significantly higher HCG on day 12 after embryo transfer than
spontaneous abortions. Only very small changes in predictive

values were observed after excluding multiple pregnancies

from the group of vital pregnancies.
As explained (Fridstrm et al., 1995), several studies have
shown that serial HCG determinations for differential diagnosis
of early pregnancy are superior to single measurements with
regard to diagnostic power (Marshall et al.,1968; Braunstein
et al., 1978; Kadar et al, 1981; Ankum et al., 1993). However,
these studies concern patients presenting with actual suspicion
of pathological pregnancy, in which active diagnostic measures
are warranted. In women who experience recurrent abortions
the most useful measurements in predicting the outcome in
early pregnancy was recently reported to be HCG and
gestation-sac diameter (Li et al., 1998). In contrast, more than
75% of the pregnant cycles in the present study led to
successful implantation. It therefore seems reasonable to spare
these patients the inconvenience of repeated blood sampling
at short intervals.
As pointed out by Glatstein et al. (1995), ideally, each
institution should analyse their own data to determine HCG
cut-offs based on their own experience. Although quality
control methods such as HCG reference preparations are
standardized between clinical laboratories, minor differences
in protocol may influence values. Glatstein et al. suggest a
prospective study in multiple centres to confirm the validity
of the cut-off values.
In summary, it can be concluded that HCG cut-off values
determined by a ROC curve analysis is useful for discriminating
between implantation successes and early pregnancy losses on
day 12 after embryo transfer. The cut-off value might aid in
the prognosis, clinical management and counselling of the
patients. The lower the HCG values found, the greater
is the need for repeated measurement and a closer follow up
of the patients. Pregnant patients with HCG values greater
than the cut-off value are followed with transvaginal US at
45 weeks post embryo transfer. Improved techniques for
identification of pathological pregnancies after IVF will be of
benefit for the patients and might also have cost saving
consequences.

References
byholm, T., Tanbo, T., Dale, P.O. et al. (1990) The first attempt at IVF
treatment. Results and requirements for satisfactory success rate. Eur. J.
Obstet. Gynecol. Reprod. Biol., 38, 125132.
Ankum, W.M., Van der Veen, F., Hamarlynck, J.V. and Lammes, F.B. (1993)
Transvaginal sonography and human chorionic gonadotrophin measurements
in suspected ectopic pregnancy: a detailed analysis of a diagnostic approach.
Hum. Reprod., 8, 13071311.
Ben-Rafael, Z., Fateh, M., Flickinger, G.L. et al. (1988) Incidence of abortion
in pregnancies after in-vitro fertilization and embryo transfer. Obstet.
Gynecol., 71, 297300.
Braunstein, G.D., Karow, W.G., Gentry, W.C. et al. (1978). First trimester
chorionic gonadotrophin diagnosis of early human pregnancy disorder. Am.
J. Obstet. Gynecol., 131, 2532.
Confino, E., Demir, R., Friberg, J. and Gleicher, N. (1986) The predictive value
of HCG- subunit levels in pregnancies achieved by in-vitro fertilization and
embryo transfer: an international collaborative study. Fertil. Steril., 45,
526531.
Correy, J.F., Watkins, R.A., Bradfield. G.F. et al. (1988) Spontaneous
pregnancies and pregnancies as a result of treatment on an in-vitro
fertilization program terminating in ectopic pregnancies or spontaneous
abortions. Fertil. Steril., 50, 8588.

1645

S.Bjercke et al.
Deutinger, J., Neumark, J., Reinthaller, A. et al. (1986) Pregnancy-specific
parameters in early pregnancies after in-vitro fertilization: a prediction of
early pregnancy loss. Fertil. Steril., 46, 7788.
Dor, J., Rudak, E., Rotmench, S. et al. (1988) The role of early postimplantation HCG concentrations in the outcome of pregnancies following
in-vitro fertilization. Hum. Reprod., 3, 663667.
Edwards, R.G., Fishel, S.B., Cohen, J. et al. (1984) Factors influencing the
success of in-vitro fertilization for alleviating human infertility. J. In Vitro
Fert. Embryo Transf., 1, 323.
Friedman, T. (1989) Infertility and assisted reproduction. Baillie`res Clinical
Obstetrics and Gynecology 1989, 3, 751766.
Fridstrm, M., Garoff, L., Sjblom, P. et al. (1995) Human chorionic
gonadotrophin patterns in early pregnancy after assisted reproduction. Acta
Obstet. Gynecol. Scand., 74, 534538.
Glatstein, I.Z., Hornstein, M.D., Kahana, M.J. et al. (1995) The predictive
value of discriminatory human chorionic gonadotrophin levels in the
diagnosis of implantation outcome in-vitro fertilization cycles. Fertil. Steril.,
63, 350356.
Hahlin, M., Sjblom, P. and Lindblom, B. (1991) Combined use of progesterone
and human chorionic gonadotrophin determination for differential diagnosis
of very early pregnancy. Fertil. Steril., 55, 492496.
Hanley, J.A. and McNeil, B.J. (1982) The meaning and use of the area under
a receiver operating characteristic curve. Radiology, 143, 2936.
Heiner, J.S., Kerin, J.F. and Schmidt, L.L. (1992) Can a single, early
quantitative human chorionic gonadotrophin measurement in an in-vitro
fertilizationgamete intraFallopian transfer program predict pregnancy
outcome? Fertil. Steril., 58, 373377.
Kadar, N., Caldwell, B.V. and Romero, R. (1981) A method of screening for
ectopic pregnancy and its indications. Obstet. Gynecol., 58, 162165.
Keith, S.C., London, S.N., Weizman, G.A. et al. (1993) Serial transvaginal
ultrasound scans and -human chorionic gonadotrophin levels in early
singleton, multiple and abnormal pregnancies. Fertil. Steril., 59, 10071010.
Li, T.C., Spring, P.G., Bygrave, C. et al. (1998) The value of biochemical
and ultrasound measurements in predicting pregnancy outcome in women
with a history of recurrent miscarriage. Hum. Reprod., 13, 35253529.
Lower, A.M. and Yovich, J.L. (1993) The value of serum levels of oestradiol,
progesteron, and beta-human chorionic gonadotrophin in the prediction of
early pregnancy loss. Hum. Reprod., 7, 711717.
Marshall, J.R., Hammond, C.B., Griff, W.C. et al. (1968) Plasma and urinary
chorionic gonadotrophin during early pregnancy. Obstet. Gynecol., 32,
760764.
McNeil, B.J., Keeler, E., Edelstein, S.J. et al. (1975) Primer on certain
elements of medical decision making. N. Engl. J. Med., 293, 211215.
Mock, P., Bischof, P., Rivest, R. et al. (1998) Modulation of human chorionic
gonadotrophin bioactivity during the first trimester of pregnancy. Hum.
Reprod., 13, 26292632.
Mol, B.W.J., van der Veen, F., Hajenius, P.J. et al. (1997). Diagnosis of
ectopic pregnancy after in-vitro fertilization and embryo transfer. Fertil.
Steril., 68, 10271032.
Shapiro, B.S., Eschbar, M., Makuech, R. et al. (1992) A model-based
prediction for transvaginal ultrasonographic identification of early
intrauterine pregnancy. Am. J. Obstet. Gynecol., 166, 14951500.
Stray-Pedersen, B. and Stray-Pedersen, S. (1984) Etiologic factors and
subsequent reproductive performance in 195 couples with prior history of
habitual abortion. Am. J. Obstet. Gynecol., 148, 140146.
Swets, J. (1979) ROC-analysis applied to the evaluation of medical image
techniques. Invest. Radiol., 14, 109121.
Tanbo, T., Omland, A., Dale, P.O. et al. (1995) In vitro fertilization/embryo
transfer in unexplained infertility and minimal peritoneal endometriosis.
Acta Obstet. Gynecol. Scand., 74, 539543.
Witt, B.R., Wolf, G.C., Wainwright, C.J. et al. (1990) Relaxin, Ca, 125,
progesteron, estradiol, Schwangerschaft protein and human chorionic
gonadotrophin as predictors of outcome in threatened and not-threatened
pregnancy. Fertil. Steril., 53, 10291036.
Yamashita, T., Okamoto, S., Thomas, A. et al. (1989) Predicting pregnancy
outcome after in-vitro fertilization and embryo transfer using estradiol,
progesterone, and human chorionic gonadotrophin -subunit. Fertil. Steril.,
59, 10811089.
Received on November 27, 1998; accepted on March 10, 1999

1646